Type 1 diabetes (T1D) results from poorly defined interactions between susceptibility genes, the environment, and the immune system. The long presymptomatic period, during which a series of islet-cell antigen-specific autoantibodies (Ab), activated T-cells and other molecular changes arise, provides an opportunity for disease prevention. Accurate risk assessment is vital for disease prevention so that therapy can be given to those individuals who are very likely to develop the disease. Despite the identification of several useful disease markers such as autoantibodies, there is still an urgent need for more specific and earlier T1D markers. Unfortunately, progress has been very slow until the recent developments of novel genomic tools. By expression profiling of peripheral blood lymphocytes, we identified several hundred genes that are differentially expressed in Ab negative (AbN) controls versus T1D patients and Ab positive (AbP) subjects. Using multivariate models of gene expression, we identified two AbP subsets who have different progression rates to T1D, suggesting that expression profiling may be very powerful T1D markers. In the R21 phase, we will carry out a cross-sectional study to confirm our preliminary findings and to develop novel biomarkers for accurate risk assessment for the transition from AbP to T1D as well as from AbN to AbP stages. Once the feasibility is demonstrated in the R21 phase, we will validate, in the R33 phase, the novel biomarkers in a prospective data set. We will also develop and validate highly reproducible and economic assays for efficient measurement of these biomarkers and the new assays will be tested in two large population screening programs (PANDA and DAISY) that have been underway in the investigators' laboratory. These step-wise studies are designed to develop and validate novel biomarkers and bring them to clinical use.